Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

iCONTRA: Toward Thematic Collection Design Via Interactive Concept Transfer

Dinh-Khoi Vo, Duy-Nam Ly, Khanh-Duy Le, Tam V. Nguyen, Minh-Triet Tran, Trung-Nghia Le

TL;DR

iCONTRA, an interactive CONcept TRAnsfer system, enables both experienced designers and novices to effortlessly explore creative design concepts and efficiently generate thematic collections, and proposes a zero-shot image editing algorithm, eliminating the need for fine-tuning models.

Abstract

Creating thematic collections in industries demands innovative designs and cohesive concepts. Designers may face challenges in maintaining thematic consistency when drawing inspiration from existing objects, landscapes, or artifacts. While AI-powered graphic design tools offer help, they often fail to generate cohesive sets based on specific thematic concepts. In response, we introduce iCONTRA, an interactive CONcept TRAnsfer system. With a user-friendly interface, iCONTRA enables both experienced designers and novices to effortlessly explore creative design concepts and efficiently generate thematic collections. We also propose a zero-shot image editing algorithm, eliminating the need for fine-tuning models, which gradually integrates information from initial objects, ensuring consistency in the generation process without influencing the background. A pilot study suggests iCONTRA's potential to reduce designers' efforts. Experimental results demonstrate its effectiveness in producing consistent and high-quality object concept transfers. iCONTRA stands as a promising tool for innovation and creative exploration in thematic collection design. The source code will be available at: https://github.com/vdkhoi20/iCONTRA.

iCONTRA: Toward Thematic Collection Design Via Interactive Concept Transfer

TL;DR

iCONTRA, an interactive CONcept TRAnsfer system, enables both experienced designers and novices to effortlessly explore creative design concepts and efficiently generate thematic collections, and proposes a zero-shot image editing algorithm, eliminating the need for fine-tuning models.

Abstract

Creating thematic collections in industries demands innovative designs and cohesive concepts. Designers may face challenges in maintaining thematic consistency when drawing inspiration from existing objects, landscapes, or artifacts. While AI-powered graphic design tools offer help, they often fail to generate cohesive sets based on specific thematic concepts. In response, we introduce iCONTRA, an interactive CONcept TRAnsfer system. With a user-friendly interface, iCONTRA enables both experienced designers and novices to effortlessly explore creative design concepts and efficiently generate thematic collections. We also propose a zero-shot image editing algorithm, eliminating the need for fine-tuning models, which gradually integrates information from initial objects, ensuring consistency in the generation process without influencing the background. A pilot study suggests iCONTRA's potential to reduce designers' efforts. Experimental results demonstrate its effectiveness in producing consistent and high-quality object concept transfers. iCONTRA stands as a promising tool for innovation and creative exploration in thematic collection design. The source code will be available at: https://github.com/vdkhoi20/iCONTRA.
Paper Structure (16 sections, 5 figures)

This paper contains 16 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Text-to-Image
  4. Text-based Image Editing
  5. Proposed System
  6. User Interface
  7. Zero-Shot Image Editing Algorithm
  8. Implementation
  9. Pilot Study
  10. Participants
  11. Baseline condition and Measurement
  12. Tasks
  13. Apparatus and Procedure
  14. Quantitative result and Qualitative feedback
  15. Limitations and Failure Cases
  16. ...and 1 more sections

Figures (5)

  • Figure 1: iCONTRA interface. a) Upload the original object image, b) Generate the Object via prompt, c) Modify the prompt until obtain the desired object.
  • Figure 2: The proposed zero-shot image editing pipeline consists of two phases: data extraction and image synthesis.
  • Figure 3: The first column represents the original objects (lamp, bag, wallet and sofa). The series of objects generated share the same concept, starting from the second column.
  • Figure 4: Quantitative data of the pilot study. Satisfying results participants generated using iCONTRA.
  • Figure 5: Failure cases which users face. A) The user desires a handbag with a yellow chain, but the model inadvertently changes the entire handbag to yellow. B) The model currently struggles to fully understand the prompt, particularly when it includes a description of a white color helmet at the end of the input text. C) The model is unable to generate different of pose or view of object. D) The model is unable to retain the original object.